Reproduce “Fast Bayesian Optimization Through Linear Models” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Colin Doumont, Henry Moss, Philipp Hennig
Fast BO reduces Bayesian optimization overhead for cheap black-box tasks. It retains an uncertainty-aware linear surrogate, simplifies fitting, and uses specialized search for Thompson sampling, upper confidence bound, and expected improvement. Across six high-dimensional benchmarks, its UCB and EI variants perform similarly to Vanilla BO with about 700 times less algorithm time, not faster evaluations or results on every problem.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.